In this paper, we introduced a lightweight object detecting system which can help to detect illegal occupation in bus lane. This system can play a role to make the city smarter. The system can classify vehicles into social cars and buses, recognize and detect car licenses. At the beginning, we introduced relevant works done by other researchers which turned out to be inspiring. Next, we designed the structure of our system into three different parts: central service module, server, and embedded device. Each part plays different roles. The central service module realizes the function of the front-end page and back-end of the platform by applying micro-service architecture. Server is responsible for works related to model training and communication between embedded device and cloud server. Embedded device should be implemented on buses, detecting, and recording illegal occupation of bus lane by running the trained object detection model. The system can realize different functions by implementing different models. We used deep learning to realize our expected function. First, a dataset includes two kinds of vehicle, car plate, 10 Arabic numbers, 26 English letters, and 6 Chinese abbreviation of 6 Chinese provincial districts. Then we use YoloV5 to train the model. After the model is trained, we evaluated the model. The results indicate that the model matches our expectation.
This work deals with the problem of high computation complexity in image registration. A hierarchical multiresolution strategy is utilized to speed up the processing of SIFT by starting on a low resolution octave. The initial affine transformation model will be achieved. In subsequent multiresolution octaves, we apply the transformation affine model getting from upper octave to current octave, then, combined with geometrical distribution of matched keypoints to further remove incorrect mappings and update affine transformation model. The strategy ends with the best affine transformation model on the bottom octave(full-size image). Experimental results show that the proposed method can achieve comparative accuracy with less computational than original SIFT
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.